Introduction: In the realm of agri-food enterprises, a perplexing paradox unfolds, captivating theattention of industry observers, researchers, and economists alike. Despite the apparentextravagance of their product pricing, these businesses not only manage to steer clear offinancial turmoil but also thrive, showcasing robust profit margins that exceed industrynorms. This phenomenon underscores the pivotal role of digital marketing in reshapingconsumer perceptions and driving demand within the agri-food sector. Central to theirsuccess is the strategic utilization of digital marketing channels to optimize profitability,despite the challenges posed by production costs. According to Järvinen & Karjaluoto [1],an organization’s attempts to implement digital marketing metrics systems, as well asthe outcomes, cannot be understood without considering the rationale for the parameterschosen, the analysis of metrics data, and the organizational circumstances surrounding thesystem’s use.Digital marketing is an effective and efficient way to promote companies online, withmany strategies and platforms accessible [2]. Vollrath & Villegas [3] suggest preventingdigital marketing analytics myopia. Marketing analytics will need to be more adept at interpreting data from a greater variety of sources as the industry transitions to a moredigital one. Even as technology and marketing channels change, practitioners and scholarsof digital marketing analytics can still create value for businesses and consumers by usingthe consumer choice journey as a strategic framework [4]. However, there is a paradox.Several businesses appear to be selling their products at high prices, and they are notbankrupting, but on the contrary, they have high profits.The scope of this article is to explore this paradoxical phenomenon. Specifically, it aims,on the one hand, to examine specific digital marketing measurement variables and howthey affect the reduction in advertising costs and, on the other hand, to propose a digitalmarketing strategy capable of reducing advertising costs to increase profitability, thuscontributing to the efficient use of resources and sustainable development. The significanceof this article lies in its potential to offer valuable insights to industry practitioners, guidingthem in the development of data-driven marketing strategies that foster sustainable growthand competitive advantage in the dynamic agri-food sector. The findings of this studycould contribute to the optimization of resource utilization by reducing advertising costsand increasing business profitability. By elucidating the impact of specific digital marketingmeasurement variables on cost reduction, this article advances the understanding andapplication of efficient strategies that promote sustainable development. Such insightsare crucial for the survival and evolution of businesses in the rapidly changing agri-foodindustry, making this research highly relevant and impactful.Relying entirely on web analytics data may result in unproductive or detrimentalmarketing decisions. As a result, businesses should only use web analytics data as a partof their performance evaluations. Web analytics may enhance healthcare websites bymonitoring engagement, users, acquisition, content, and platforms, thereby improvingusability and conversion rates [5]. Improving web analytics systems may increase anorganization’s profitability by assessing user behavior, performance results, and satisfactionof consumers [6]. Key parameters of digital marketing, extensively researched in thiswork, include social and search traffic sources, the abandonment rate, the pages per visit,and the time customers spend on their websites, along with the number of returningwebsite customers.By harnessing the power of digital marketing metrics, agri-food enterprises can effec-tively reduce advertising costs, achieving profitability even in the face of high productioncosts. This strategic alignment with digital marketing principles empowers them to not onlysurvive but thrive in today’s competitive landscape, forging a path to sustained success inthe agri-food industry. Therefore, the following research question is raised:RQ: “How are specific digital marketing measurement variables aligned with the reduction inadvertising costs?”The research question aims to determine what the best channel for digital marketingis in order to minimize the advertising costs and increase the profitability of agri-foodbusinesses. Directing resources to the lowest-cost choice of digital advertising channels hasas the result of, in addition to increasing profits, the saving of resources, thus contributingto sustainable development. Our research has analyzed digital marketing variables andfocuses on social and search sources variables, suggesting a digital marketing strategy ofdirecting resources to search sources rather than social sources.The novelty of this study lies in the fact that it attempts to investigate how, through theselection of an appropriate digital marketing strategy, companies can reduce advertisingcosts and therefore increase their profitability, thus making efficient use of resources andcontributing to sustainable development. It was revealed by our bibliographic analysisthat there is a dearth of research that is comparable to this one that concurrently looks atsustainable development, business profitability, and digital marketing strategies.This article makes a significant contribution to understanding how digital marketingstrategies can optimize profitability and resource efficiency in the agri-food industry. By ex-ploring the relationship between digital marketing variables and advertising costs, through robust analytical methods like regression analysis and fuzzy cognitive mapping, this studyreveals actionable insights for agricultural enterprises. It underscores the importance ofstrategic digital investments, particularly in search sources, to reduce advertising costs.Moreover, the findings highlight a shift towards sustainability by minimizing resourcewaste, thereby setting a precedent for enhancing competitiveness and operational efficiencyin the sector.Digital marketing strategies may include metrics such us website traffic, user engage-ment, conversion rates, customer demographics, and purchasing behavior. By analyz-ing these metrics, agri-food companies can gain valuable insights into consumer prefer-ences, market trends, and the efficacy of their marketing campaigns. Ghahremani-Nahr &Nozari [7] claim that key performance indicators are critical and delicate indicators for anyfirm that can successfully identify and regulate them. Measuring the effectiveness of digitalmarketing activities and using important indications of digital marketing performancehelps boost marketing productivity while also improving the efficacy and optimization ofmarketing expenditure budgets. There are several measures for measuring digital market-ing performance that can help improve the effectiveness of marketing initiatives. Moreover,the concept of “efficient promotions” involves the ability to allocate resources wisely andemploy marketing tactics that yield the highest return on investment (ROI). It entails cus-tomizing advertising activities to resonate with target audiences, optimizing advertisingexpenditure, and maximizing conversion rates.Our research underscores its originality by focusing on the intersection of digitalmarketing strategies in the agri-food sector, particularly highlighting the innovative useof website customer behavioral data in examining the impact of traffic sources (searchand social) and other metrics on their advertising costs. Through the methodology ofstatic FCM simulation, a unique lens is provided for exploring how optimizing digitalmarketing expenditures, specifically between social media and search engine resources [8],can enhance both profitability and sustainability within agri-food businesses. This workrepresents a significant contribution to the literature as it highlights the importance andimplication of digital marketing strategies in the agri-food sector. It focuses on cost reduc-tion through optimizing advertisements on social media and efficiently using resources forsearch engine optimization. The reduction in these expenses has substantial implicationsnot only for the profitability of businesses but also for sustainability, ensuring efficient useof limited natural resources and long-term viability of the sector. In this way, this researchdelves into identifying how website customer behavior can determine beneficial digitalmarketing strategies in the agri-food sector, and the firms’ sustainability and resourceefficiency. Therefore, in Table 1the findings of recent and relevant studies in the field ofagri-food and agricultural sectors is laid out to highlight the innovative implications ofour research.In the following paragraphs of Section 2, a presentation of the literature regarding thepeculiarities of the agri-food sector, the economic dimension of production costs in agri-foodbusinesses, and the contribution of digital marketing and big data are provided. Overall,the literature emphasizes the importance of understanding the distinct characteristics of theagri-food sector, effectively managing production costs, and leveraging digital marketingand big data analytics to address the phenomenon, as companies in the sector can overselltheir products. The development of research hypotheses then follows, as well as theresearch methodology. In Section 3, the main findings, from the statistical analysis of thedata and the simulation of five basic scenarios of changing the primary parameters asshown by the statistical processing of the data, are presented. In Section 4, a discussionis held where the main findings of the statistical analysis and the results of the scenariosimulation are interpreted, highlighting the proposed digital marketing strategy. Finally,Section 5briefly presents the conclusions of this research.2. Materials and Methods2.1. Peculiarities of the Agri-Food SectorThe agri-food sector encompasses a diverse range of activities, including agriculture,food processing, distribution, and retailing. This sector is characterized by unique featuressuch as seasonality, perishability of goods, and dependency on environmental factors.Researchers have extensively studied these peculiarities to understand their implicationsfor market dynamics, supply chain management, and policy development. Shumakovaet al. [16] identify the peculiarities of agricultural production including reliance on naturaland climatic conditions, low monopolization of food producers, high capital-output ratio in iry cattle breeding, low profitability, cost disparities, and inadequate social environmentsin rural areas. The identification of trends and possibilities for strategic management willaid in promoting the process of improving relationships between various kinds of marketinteraction and mitigating the impact of risks on the agri-food market. Melikyan [17]concluded that the complex’s financial and economic decisions serve the interests of bothagrarian and industrial groups, rather than those of just one institution.Furthermore, corporate profits are consolidated through incomplete compromisesbased on interest in the food industry rather than market transfer rates. This allows profitgeneration capabilities to be transmitted between intra-enterprise organizations. To createa favorable investment climate in the food industry, administrative barriers to commercialrelations (agricultural, processing, and commerce) should be minimized through corporategovernance measures. The global commerce in agri-food items, particularly grains andsunflower oil, is mostly focused in three regions: Europe and North and South America. Atthe same time, exports from Asia, South and Central America, and Africa are expanding themost rapidly. Considering the development of international economic integration, it shouldbe noted that the main suppliers of agri-food products are mainly the most developed inte-gration groups—the EU, NAFTA, and Southern Common Market(MERCOSUR)—wherethe largest share belongs to the leading countries. Integration groups have been urged tocontribute to the solution of the food security challenge at the regional and internationallevels [18].The agri-food sector is crucial to the national economies of European Union (EU) Mem-ber States. However, despite support from the first pillar of the CAP (Common AgriculturalPolicy of the EU), agricultural sectors face long-term challenges such as low food quality,insufficient knowledge of legal documents on environmental, land, and food law, andlimited market access for farmers and producers [19]. Busch [20] examines the evolution ofthe agricultural sector through the sociological lens of neoliberalism’s conflict. He arguesthat the theory of neoliberalism, as applied to the agricultural sector, will face a series ofobstacles: climate change, rising energy costs, reduction in biodiversity, environmentaldegradation due to reckless water and soil usage, and the pursuit of sustainability.Agri-food value chains play a significant role in shaping the labor dynamics that existtoday in the industry, and the sustainability of these value chains is highly dependenton good working conditions [21]. With greater participation in agricultural global valuechains leading to increased agricultural employment growth, primarily driven by theprocessed food sector downstream, and with more pronounced effects in lower-middle-income and high-income countries, the rise of agricultural global value chains (AGVCs)has fundamentally changed the nature of food production worldwide [22]. By integratingsustainability drivers into value-chain governance, smallholders in developing nations cannow access higher-value markets through a comprehensive approach that strikes a balancebetween profit maximization and social and environmental impacts [23,24].However, as per Meemken et al. [25], sustainability standards can enhance productionprocedures in certain instances, but they fall short on guaranteeing food system sustain-ability on a large scale or achieving equity goals in agri-food supply chains. Furthermore,businesses struggle to integrate sustainability in global agri-food supply chains due to adouble-company lens and information asymmetry between companies and consumers;rather than focusing on major operations and supply chain improvements, rational busi-nesses instead tend to focus on symbolic actions and communication campaigns usingbrand-enhancing marketing tools like sustainability reports [26]. According to Struik andKuyper [27], to make agriculture “green” once more while balancing resource-use efficiencywith social, economic, and environmental aspects, sustainable intensification in agriculturerequires precise principles, cost–benefit analysis, and common norms. Circular use ofbiotic resources and avoiding losses are key to enhancing efficiency in agri-food systems,ensuring food security, and the functioning of the Earth’s system [28]. Sustainability 2024,16, 5889 6 of 252.2. Financial Dimension of the Agri-Food SectorOne of the critical economic dimensions of the agri-food sector is the cost of production.Production costs in agriculture and food processing businesses significantly impact theircompetitiveness, profitability, and sustainability. Studies have analyzed various factorsinfluencing production costs, including input prices, technology adoption, labor availability,and regulatory compliance. As noted by Reardon [29], the midstream segment has expe-rienced remarkable expansion and transition. This includes both a “modern revolution”with huge multinational firms and a “silent revolution” with a proliferation of small andmedium enterprises and significant investment. Revolutions have been fueled by directgovernment action, followed by liberalization, privatization, urbanization, income growth,and infrastructure improvements. Private-sector investment has been overwhelminglyimportant compared to direct government investments and for the domestic market, com-pared to internationally traded sectors. Agri-food policy had a significant part in theseprocesses. Sometimes, the roles have been slowing transforming. Policy and public in-vestment have played a significant role in transforming the food value chain. Investmentsin production technical modernization have a substantial impact on the competitivenessof the national agri-food system [30]. Enabling conditions, such as roads and electricity,can increase the profitability of private investments by boosting supply from upstream tomidstream. By combining technical breakthroughs with sociocultural and policy reforms,agri-food systems can be transformed to meet climate, economic, environmental, health,and social concerns [31].A point that deserves attention is the financialization of the agri-food sector. Financialinvestment in the food and agriculture sectors has increased in recent decades, especially inequity-related funds that invest in or follow the performance of a variety of publicly tradedinternational agri-food firms. At their peak in recent years, equity-related investmentfunds accounted for around one-third of total financial investment in the sector. Largeasset management firms have incentives to encourage agribusiness firms in which theyown a significant stake to pursue strategies that result in higher returns to shareholders,particularly market strategies that benefit the entire sector rather than individual firms.Such incentives can lead to anti-competitive behavior, mergers and acquisitions, andincreased entry barriers into specific industries, resulting in price increases [32].Thus, examining the extent to which the agricultural sector is integrated into the stockmarket is important for a comprehensive understanding of the causes and consequences ofincreasing prices of agricultural products. When there is a financial crisis, managers oftenturn to the cost-cutting approach, with advertising costs usually affected first. The results ofthe Markota et al. [33] survey show that firm profitability and advertising expenditures arepositively correlated, but firm size and legal structure do not have a statistically significanteffect on the level of advertising expenditures. Thus, in agri-food sector businesses, as inall businesses, the reduction in advertising costs is an important factor in the profitabilityof businesses. Decision makers should evaluate which parameter has the most impacton advertising costs and then compare whether reducing or increasing spending on thisparameter ultimately benefits the company’s profitability.According to Coetzee, K. [34], sustainable farming is commonly associated withenvironmental concerns, but the significance of profitability in achieving sustainabilityis often overlooked. Profitability is a critical factor that should not be ignored whendiscussing sustainable farming practices, as it is essential for the long-term success andviability of agricultural operations. Ensuring profitability in sustainable farming practicesallows farmers to invest in better technologies and methods that can further enhance bothproductivity and environmental stewardship. Thus, integrating digital marketing strategiesin agriculture can play a pivotal role in promoting both profitability and sustainability,ensuring the sector’s growth and resilience in the global market. Sustainability 2024,16, 5889 7 of 252.3. Contribution of Big Data and Digital MarketingThe emergence of digital marketing and big data analytics has transformed the agri-food industry’s marketing practices. Digital marketing, big data, and web analytics allplay an important role in the growth and longevity of a company’s digital brand, as wellas profitability [35]. Digital technologies enable businesses to target consumers moreeffectively, optimize marketing strategies, and enhance customer engagement. Big dataanalytics offer valuable insights into consumer preferences, market trends, and supply chainefficiency, empowering agri-food businesses to make data-driven decisions. According toRavi & Rajasekaran [36], the concept of “digitalization” has started taking over the globe,and with it, a wide range of digital marketing innovations and techniques can be appliedmore efficiently to enhance the conventional marketing approach. Using digital marketingtools is one of the best ways to reach and draw in customers.Digitalization opens up new opportunities for agriculture and the agro-industrial com-plex. Digital platforms and big data can be used for several operations in the agro-industrialcomplex, including planning, production, sales, financial management, personnel, andtechnical advancements. Digital platforms and big data can enhance national food securityby improving management, decision-making, product quality, production costs, market-ing, and agricultural raw-material processing precision [37]. According to Tatkari [38],e-commerce in agriculture refers to the purchase and sale of agricultural goods and servicesvia electronic methods. Online marketplaces, mobile apps, and other digital platforms canhelp expedite transactions between farmers, buyers, and stakeholders in the agriculturevalue chain. E-commerce in agriculture opens up new markets and clients outside tradi-tional borders. This can help farmers reach a larger client base, boost sales, and minimizereliance on local markets. E-commerce platforms can also give farmers access to real-timemarket information, allowing them to make better pricing and production decisions. As anillustration, e-commerce adoption has boosted agricultural production efficiency in China,supporting rational labor, land, and capital allocation while also contributing to sustainablegrowth and a sense of community [39].Another advantage of e-commerce in agriculture is its capacity to simplify the supplychain and lower transaction costs. Online marketplaces enable farmers to connect withcustomers directly, eliminating the need for intermediaries and increasing supply chainefficiency. Lower transaction costs benefit both buyers and sellers, making it easier forfarmers to sell their products and purchasers to purchase them. As food security is underthreat, society must use creative technology solutions to secure a healthy and safe foodsupply while lowering the negative environmental implications of agricultural production.This can only be accomplished by ensuring that agricultural technologies and productsare consistent with societal expectations, needs, and priorities. Digital technology andbig data tools can promote agricultural innovation by increasing production efficiencyand addressing societal concerns, but their acceptance is dependent on technological andeconomic variables [40].While the general notion that consumers are “anti-agri-food technology” is rejected, itis important to co-develop applications of agri-food technologies with stakeholders andother end-users, including consumers [41]. Genetic modification, food fortification, andprocessing technologies are all gaining popularity as potential solutions to future foodsecurity and safety concerns. Consumers and other participants in the food supply chainmust accept such technologies to ensure their successful adoption. Basso & Antle [42] claimthat the global food system needs to become more sustainable. Thus, digital agriculture,which uses digital and geospatial technology to monitor, assess, and manage soil, climate,and genetic resources, demonstrates how to tackle this issue while balancing the economic,environmental, and social aspects of sustainable food production.In terms of customer evaluation, major drivers include trust in institutions, informationassessment, perceived risks and advantages, attitudes toward the product or technology,perceived behavioral control, product quality perception, and health impact [43]. Thisis where the role of digital marketing comes into play. Through appropriate strategies, Sustainability 2024,16, 5889 8 of 25marketing managers should make consumers aware of technological advancements in theagri-food sector and influence their psychological perspective so that they accept thesechanges and perceive them as added value to the product they are purchasing.The use of digital tools such as e-commerce, social networks, search engine optimiza-tion (SEO), mobile applications, and others have demonstrated their high performance andleadership role in the rapid processing of market data, building effective communications,lowering costs, and mastering new consumer segments. The world’s leading marketersinsist on these processes, which are described in modern terms such as “marketing 4.0”and “marketing 5.0” [44]. Farmers can use digital marketing to boost their product’s sellingprice and lower their marketing expenses. Young farmers in particular are prepared touse digital marketing. To build a sustainable digital agriculture market, the federal andstate governments can conduct programs to teach farmers about digital marketing [45].According to Yekimov et al. [46], agricultural businesses can become more competitive byusing social networks to promote their products and services.However, for this process to be successfully organized and carried out, experts withthe necessary training and expertise in social media product and service promotion mustbe used. When utilizing social media for marketing purposes, we should not anticipateimmediate results; nevertheless, persistent efforts in this regard can draw in a sizablenumber of prospective customers. Dollega et al. [47] found that social media can be usefulin part because it increases web traffic but only slightly boosts product orders and sales;in contrast, the more extensive social media campaigns produce a significantly greaternumber of orders on Facebook, the most popular social media platform, surpassing thoseon less-popular ones like Instagram.Launching into this exploration of digital marketing tools, it is imperative to addressManko’s [48] research; rather than using traditional tactics like printing paper vouchers,cards for punching, and leaflets to promote special bargains, customers may easily accesstheir rewards through a personalized mobile app. This application is a comprehensivesolution that facilitates customer recruitment, offer dissemination, communication of pro-motions, order placement, introduction of new products and services to existing customers,and a variety of other activities. Leveraging app development for digital marketing hasthe potential to greatly increase sales by better engaging the target audience and providingvital business insights via user analytics.2.4. Research HypothesesAfter studying the relevant and recent scientific literature, this research intends toanswer the following research hypotheses, based on the main research question out-lined above.Hypothesis 1: The advertising costs of agri-food firms are related to social and search traffic sources.This hypothesis suggests that the amount of advertising expenditure undertaken bycompanies in the agri-food sector is correlated with the traffic originating from social mediaplatforms and internet search sources. This implies that companies investing more inadvertisements on these platforms are likely to incur higher advertising costs. Conversely,a decrease in advertising expenditure on these traffic sources could be associated with areduction in the overall advertising costs of agri-food firms.Hypothesis 2: The advertising costs of agri-food firms are related to the abandonment rate oftheir websites.This hypothesis suggests that there exists a relationship between the advertisingexpenses incurred by agri-food companies and the rate at which visitors abandon theirwebsites without completing desired actions. A higher abandonment rate may indicateinefficiencies or inadequacies in the website design, content, or user experience, whichcould necessitate increased advertising efforts to attract and retain visitors. Conversely, Sustainability 2024,16, 5889 9 of 25lower abandonment rates may correlate with more effective website optimization strategies,potentially resulting in reduced advertising costs for agri-food firms.Hypothesis 3: The advertising costs of agri-food firms are related to the pages per visit and timeon site customers spend on their websites.This hypothesis proposes a connection between the advertising expenses of agri-food firms and two metrics: the average number of pages visited per session and theduration of time customers spend on their websites. A higher number of pages per visitand longer time spent on the site may suggest greater engagement and interest amongvisitors, potentially leading to increased brand awareness and higher conversion rates.Consequently, agri-food companies may need to allocate more resources to advertisingactivities to maintain and enhance this level of engagement. Conversely, lower engagementmetrics may indicate a need for more targeted advertising efforts or website improvements,which could potentially result in reduced advertising costs.Hypothesis 4: The advertising costs of agri-food firms are related to the number of their returningwebsite customers.This hypothesis posits a correlation between the advertising expenditures of agri-foodcompanies and the number of customers who revisit their websites. A higher number ofreturning website customers may indicate stronger brand loyalty and satisfaction with theproducts or services offered by agri-food firms. Consequently, these companies may needto invest more in advertising activities to maintain and further cultivate customer loyalty.Conversely, a lower number of returning website customers may suggest the need for moretargeted advertising efforts or improvements in customer retention strategies, potentiallyleading to reduced advertising costs.2.5. Methodological ConceptThe primary objective of this study is to determine the relationship between digitalmarketing metrics and the advertising costs (both organic and paid traffic costs) incurredby agri-food companies, using a three-phase structure. These variables are common digitalmarketing KPIs [49]. The methodological context refers to discrete stages, such as thecollection of big data, statistical analyses, and dynamic modeling.The first stage of the research context focuses on the extraction of the website analyticaldata from the selected organizations for this study. These data were extracted from fiveof the biggest firms in the agri-food sector, by utilizing the Semrush [50] website software.These data were later used as input for the upcoming statistical analysis.In the second stage, the retrieved data were harvested in the execution of statisticaltests, like descriptive statistics, correlation, and linear regression analyses. Through theextracted coefficients, the impact of social and search source traffic, website bounce rate,pages per visit, and time on site, as well as returning customers, on agri-food advertisingcosts were examined. Moreover, the research hypotheses were verified.The authors proceeded to the next step of the methodological context after extractingthe regression and correlation coefficients, using a fuzzy cognitive map (FCM) model tovisualize the relationships between the variables and run simulation scenarios. Fuzzy cogni-tive maps are popularly used for knowledge representation and reasoning in a wide rangeof application domains and have attracted a lot of research interest. They are useful for avariety of tasks, including forecasting, analysis, modeling, and decision-making [51]. Fuzzycognitive maps can be efficiently analyzed and designed for practical uses by breakingthem down into fundamental modules and studying their inference patterns hierarchi-cally [52]. The efficacy of digital product and service internet marketing strategies isefficiently assessed by the fuzzy cognitive mapping method [53]. With online direct mar-keting applications, the fuzzy cognitive mapping method helps to improve the efficacyof these efforts by evaluating negative factors such as viruses and spam [54]. The fuzzy Sustainability 2024,16, 5889 10 of 25cognitive map offers a well-informed framework for an objective analysis of the dynamicsof digital entrepreneurship by efficiently identifying and analyzing its determinants [55].By adding time and offering dynamic scenario changes, fuzzy cognitive maps can improvestrategy maps and enable hierarchical performance measurement hierarchies [56]. Modelscreated by recurrent artificial neural networks, like fuzzy cognitive maps, are collections ofconcepts or neurons and the various causal connections that exist between them. Usersshould select the best kind of FCM based on the following factors: (a) the nature of theproblem; (b) the problem’s necessary representation capabilities; and (c) the degree ofinference required by the case [57].2.5.1. Sample DescriptionIn this stage of the study, the selection of the sample is presented, based on the selectedagri-food companies that were included in this study. These refer to the five biggest agri-food firms for 2023, as they were categorized by Globaldata [58], based on their marketcapitalization in 2023. Thus, Nestle SA, Mondelez International Inc., The Kraft Heinz Co.,Danone SA, and The Hershey Co. make up the sample for the present research. To performweb analytics collection from the corporate websites, the authors utilized the web platformdecision support system (DSS) from Semrush [50]. Data gathering involved 180 days ofobservation and extraction, from July 1st, 2023, to January 31st, 2024. The collected analyticmetrics are presented in detail in Table 2below.Table 2. Description of collected web analytics.Analytics/Metrics DescriptionAdvertising CostsAdvertising costs are all the costs and expenditure associated with marketing and promotion, including,without limitation, advertising, agency fees, materials, medical affairs, meetings, and, when notspecifically excluded, allocated sales force costs. In this research, advertising costs consist of organic andpaid campaign costs [59].Direct SourcesDirect sources refer to the traffic or visitors that arrive at a website directly, without the use ofintermediary sources such as search engines, links from other websites, or social media. Examplesinclude users typing the URL into their browser, using bookmarks, or access via saved links.Referral SourcesReferral sources represent the websites or platforms that drive traffic to the site through links, such asarticles, blogs, forums, and communities, targeting audiences already interested in related contentor services.Social Sources Social sources refer to the traffic or visitors generated from social media platforms.Search SourcesSearch sources refer to the traffic sources of a website originating from search engines like Google, Bing,Yahoo, and others. These sources track the traffic generated from search engine results, typically whenusers are actively seeking specific content or information.Bounce Rate The percentage of visitors who navigate away from a website after viewing only one page indicates alack of engagement.Pages per Visit The average number of pages a visitor views during a single session on a website indicates the level ofexploration and engagement.Time on Site The average amount of time visitors spend on a website during a single session provides insights intouser engagement and interest.New Customers A new customer refers to a person or organization that was recently acquired via online channels.Returning CustomersReturning customers are individuals or entities who have previously interacted with a business or brandonline and have subsequently returned to engage in further transactions or make additional purchases.2.5.2. Conceptual FrameworkFuzzy logic is used to represent all variables and metrics in the conceptual model,Cognitive Mapping (FCM) DSS software of MentalModeler [60]. This method shows howchanges to one metric may affect others by highlighting the relationships and connectionsbetween this study’s variables [61] while creating a .mmp file for future analyses and Sustainability 2024,16, 5889 11 of 25scenario development. FCM might be able to be arduously employed as a stationary modelfor simulation. To further tackle the task of changes in a metric, subjective fuzzy cognitiveapproaches are used for the defuzzification process, converting knowledge from feedbackinto statistical information [62]. The use of the fuzzy cognitive macro-scale framework inthis paper is demonstrated in Figure 1, showing the connections between all the variablesand factors that were looked at. The relationships between the variables, as depicted intheir affiliation statistics, encompass both direct and inverse correlations, stemming fromcorrelation analyses. A blue line represents a positive relationship, and an orange linerepresents a negative relationship. Additionally, the line’s width intensifies in proportionto the strength of the correlation.Sustainability 2024, 16, 5889 12 of 26 Figure 1. Fuzzy cognitive mapping model. Blue and red arrows signify positive and negative corre-lations between variables, respectively. The symbols “+” and “–” represent the positive and negative percentage changes, respectively. 3. Results 3.1. Statistical Analysis After defining this study’s sample and research design, the authors carried out the necessary statistical analysis to extract the relevant coefficients from the relationships between the variables. Initially, Table 3 displays the fundamental descriptive statistics of the independent and dependent variables. In Table 4, the correlations of this study’s Figure 1. Fuzzy cognitive mapping model. Blue and red arrows signify positive and negativecorrelations between variables, respectively. The symbols “+” and “–” represent the positive andnegative percentage changes, respectively. Sustainability 2024,16, 5889 12 of 253. Results3.1. Statistical AnalysisAfter defining this study’s sample and research design, the authors carried out thenecessary statistical analysis to extract the relevant coefficients from the relationshipsbetween the variables. Initially, Table 3displays the fundamental descriptive statisticsof the independent and dependent variables. In Table 4, the correlations of this study’svariables are presented. To support the creation of the advertising costs variable (consistingof organic and paid traffic costs), KMO and Cronbach’s alpha values were higher than0.7 [63,64], meaning that the variable is proper for statistical analysis and is cohesive(Table 5).Table 3. Descriptive statistics.Mean Min Max Std. Deviation Skewness KurtosisAdvertising Costs 246,125.66 147,070.00 426,498,00 89,226.31 1.170 0.716Direct Sources 323,284.57 263,604.00 411,527.00 53,683.07 0.636 −0.663Referral Sources 373,087.43 265,622.00 552,072.00 88,429.30 1.463 1.911Social Sources 5985.14 2431.00 10,992.00 2996.30 0.730 −0.193Search Sources 147,035.29 96,976.00 193,138.00 32,360.16 −0.173 −0.514Bounce Rate 0.53 0.49 0.57 0.03 0.143 −1.717Pages per Visit 2.75 2.62 2.85 0.09 −0.246 −1.957Time on Site 500.14 370.00 691.00 114.01 0.764 −0.253New Customers 285,612.00 248,488.00 338,317.00 36,169.57 0.613 −1.458Returning Customers 849,392.71 698,598.00 106,4952.00 130,208.89 0.534 −0.360Table 4. Correlation analysis.Advertising CostsDirect SourcesReferral SourcesSocial Sources[Search SourcesBounce RatePages per VisitTime on SiteNew CustomersOld CustomersAdvertising Costs 1 0.097 −0.049 0.368 −0.130 −0.149 −0.313 0.328 0.250 −0.017Direct Sources 0.097 1 0.430 0.223 −0.126 0.292 0.753 −0.225 0.699 0.678Referral Sources −0.049 0.430 1 −0.433 0.290 0.615 0.379 0.376 0.758 * 0.919 **Social Sources 0.368 0.223 −0.433 1 0.255 −0.618 −0.017 0.027 0.239 −0.116Search Sources −0.130 −0.126 0.290 0.255 1 −0.487 0.208 0.213 0.364 0.399Bounce Rate −0.149 0.292 0.615 −0.618 −0.487 1 0.003 0.223 0.310 0.403Pages per Visit −0.313 0.753 * 0.379 −0.017 0.208 0.003 1 −0.565 0.409 0.619Time on Site 0.328 −0.225 0.376 0.027 0.213 0.223 −0.565 1 0.357 0.216New Customers 0.250 0.699 0.758 * 0.239 0.364 0.310 0.409 0.357 1 0.899 **Returning Customers −0.017 0.678 0.919 ** −0.116 0.399 0.403 0.619 0.216 0.899 ** 1*, ** indicate statistical significance at the 95% and 99% levels, respectively.Table 5. Consistency of the advertising costs variable.Cronbach’s Alpha Kaiser–Meyer–Olkin Factor AdequacyAdvertising Costs (Organic andPaid Traffic Costs) 0.781 0.796The simple linear regression (SLR) models that were developed aimed to highlight thestatistical significance of the study variables’ relationships (Table 6). As a dependent variable,the advertising costs of the sample’s agri-food firms were selected, and the web analyticsof social and search sources, bounce rate, returning customers, pages per visit, and timeon site were selected. So, for the first SLR model of advertising costs with the independentvariables of social and search sources, the model’s variables were overall verified, withp-values < a = 0.05level of significance and R2= 0.689. When social and search sourcesincrease by 1%, agri-food firms’ advertising costs variate by 28.8%, and−17.7%, respectively. Sustainability 2024,16, 5889 13 of 25Table 6. Impact of social and search sources traffic on agri-food firms’ advertising costs.Variables Standardized Coefficient R2Fp-Value D-W statSocial Sources 0.288 0.689 1.983 0.035 * 1.060Search Sources −0.177 0.041 ** indicates statistical significance at the 95% level.The examination of agri-food firms’ website bounce rate and returning customers ontheir advertising costs is shown in Table 7. There, the verification of the linear regressionmodel can be discerned, with p-values < a = 0.05 level of significance and R2= 0.708. Forevery 1% increase in the bounce rate and returning customers, agri-food firms’ advertisingcosts vary by−255.4% and−376.7%, respectively. As for the linear regression model of theadvertising costs with independent variables of the pages per visit and time on site, it isalso verified overall, with p-values < a = 0.05 level of significance and R2= 0.632 (Table 8).By each 1% increase of pages per visit and time on site, agri-food advertising costs variateby −284.8% and −109.7%, respectively.Table 7. Impact of website bounce rate and returning customers to agri-food firms’ advertising costs.Variables Standardized Coefficient R2Fp-Value D-W statBounce Rate −2.554 0.708 2.086 0.049 * 1.128Returning Customers −3.767 0.021 * 0.994* indicates statistical significance at the 95% level.Table 8. Impact of pages per visit and time on site to agri-food firms’ advertising costs.Variables Standardized Coefficient R2Fp-Value D-W statPages per Visit −2.848 0.632 1.791 0.025 * 1.262Time on Site −1.097 0.047 * 0.989* indicates statistical significance at the 95% level.3.2. Fuzzy Cognitive Modeling ScenariosFuzzy Cognitive Modeling (FCM) significantly enhances the examination of digitalmarketing performance in the agri-food sector by handling the complexity and uncer-tainty inherent in this field. FCM models the intricate interdependencies between variousfactors such as market trends, consumer behavior, and supply chain logistics, providingdynamic scenario analysis and robust decision support [65]. By integrating both qualitativeinsights and quantitative data, FCM offers a comprehensive and holistic view of marketingperformance [66], identifying key performance indicators and facilitating continuous im-provement. Its visual and adaptable nature allows stakeholders to intuitively understandcomplex interactions and optimize digital marketing strategies effectively, ensuring betterresource allocation and strategic planning in a rapidly changing market environment. Re-garding the wider food sector, Sarkar et al. [67] utilized a FCM simulation model to analyzethe feasibility of food storage, while Emir & Ekici [68] showed that through FCM models,the right polices can be adopted to reduce food waste.Afterward, a simulation is run, focusing on social and search sources, involving thefollowing five scenarios. The variable “social sources” is increased by 100% in the firstscenario. In the second case, there is a 100% decrease in social sources. The third scenariosaw a 100% increase in the search sources variable, while the fourth scenario saw a 100%decrease in search sources. In the final scenario, both variables undergo a combination ofchanges, resulting in a 100% decrease in the social sources and a 100% increase in the searchsources variable. The rest of the variables do not change during the process. Figure 2a–erepresent the impact of the selected variations of social and search traffic sources to variousdigital marketing KPIs of agri-food firms. Sustainability 2024,16, 5889 14 of 25Sustainability 2024, 16, 5889 15 of 26 2a–e represent the impact of the selected variations of social and search traffic sources to various digital marketing KPIs of agri-food firms. (a) (b) Figure 2. Cont. Sustainability 2024,16, 5889 15 of 25Sustainability 2024, 16, 5889 16 of 26 (c) (d) Figure 2. Cont. Sustainability 2024,16, 5889 16 of 25Sustainability 2024, 16, 5889 17 of 26 (e) Figure 2. (a) Impact of the increase in the social sources variable by 100%. (b) Impact of the decrease in the social sources variable by 100%. (c) Impact of the increase in the search sources variable by 100%. (d) Impact of the decrease in the search sources variable by 100%. (e) Impact of the increase in the search sources variable by 100% and the reduction in the social sources variable by 100%. 3.2.1. First Scenario: Increase the Social Sources Variable by 100% In the first scenario (Figure 2a), the social resources variable increases steadily by 100%, resulting in the bounce rate variable decreasing by 10% and the referral sources variable decreasing by 3%. Also, the search sources variable increases by 4%, and the direct sources variable increases by 1%, increasing the promotion cost variable by 6%. 3.2.2. Second Scenario: Decrease the Social Sources Variable by 100% Then, in the second scenario (Figure 2b), the social sources variable is reduced by 100%. The result is that the search sources variable is reduced by 10% and the direct sources variable is reduced by 3%. While the referral sources variable increases by 5%, the bounce rate variable increases spectacularly by 16%. The result of all these changes is that there is a significant reduction in the promotion cost variable by 13%. 3.2.3. Third Scenario: Increase the Search Sources Variable by 100% Afterwards, the third scenario (Figure 2c) of the simulation is carried out, in which the search sources variable increases by 100%. The result is that the variables referral sources, pages per visit, returning customers, and time on site increase by one percent, respectively, while the social sources variable increases by 2%. Also, the bounce rate variable decreases by 4%, with the result that all these changes lead to a minimum reduction in the advertising cost variable by 1%. Figure 2. (a) Impact of the increase in the social sources variable by 100%. (b) Impact of the decreasein the social sources variable by 100%. (c) Impact of the increase in the search sources variable by100%. (d) Impact of the decrease in the search sources variable by 100%. (e) Impact of the increase inthe search sources variable by 100% and the reduction in the social sources variable by 100%.3.2.1. First Scenario: Increase the Social Sources Variable by 100%In the first scenario (Figure 2a), the social resources variable increases steadily by 100%,resulting in the bounce rate variable decreasing by 10% and the referral sources variabledecreasing by 3%. Also, the search sources variable increases by 4%, and the direct sourcesvariable increases by 1%, increasing the promotion cost variable by 6%.3.2.2. Second Scenario: Decrease the Social Sources Variable by 100%Then, in the second scenario (Figure 2b), the social sources variable is reduced by100%. The result is that the search sources variable is reduced by 10% and the direct sourcesvariable is reduced by 3%. While the referral sources variable increases by 5%, the bouncerate variable increases spectacularly by 16%. The result of all these changes is that there is asignificant reduction in the promotion cost variable by 13%.3.2.3. Third Scenario: Increase the Search Sources Variable by 100%Afterwards, the third scenario (Figure 2c) of the simulation is carried out, in which thesearch sources variable increases by 100%. The result is that the variables referral sources,pages per visit, returning customers, and time on site increase by one percent, respectively,while the social sources variable increases by 2%. Also, the bounce rate variable decreasesby 4%, with the result that all these changes lead to a minimum reduction in the advertisingcost variable by 1%. Sustainability 2024,16, 5889 17 of 253.2.4. Fourth Scenario: Decrease the Search Sources Variable by 100%Then, the fourth scenario is run (Figure 2d), which decrements the search sourcesvariable by 100%. It is observed that the referral sources variable decreases by 5%, the socialsources variable decreases by 11%, the variables pages per visit and returning customersdecreases by 6%, and the time on site variable was reduced by 8%. The bounce rate variableis increases by 15%, and the direct sources variable increases by 2%. The result is that thereis an increase in the advertising cost variable by 4%.3.2.5. Fifth Scenario: Increase the Search Sources Variable by 100% and Decrease the SocialSources Variable by 100%Finally, the 5th scenario (Figure 2e) is carried out where a simultaneous change in thesocial sources variable is attempted, a decrease of 100%, and the search sources variableincreases by 100%. In this case, it is observed that the referral sources variable increases by5%, and the bounce rate variable increases by 14%. The pages per visit variable increases by1% as well as the returning customers variable, and the time on site variable increases by2%, while the direct sources variable decreases by 4%. The result is that there is a significantreduction in the advertising cost variable by 14%. This reduction in the advertising costvariable is the maximum observed in all scenarios.4. DiscussionThe analysis reveals diverse distribution patterns across the variables (Table 3). Skew-ness, indicating the asymmetry of the distribution, unveils interesting insights. Variablessuch as advertising costs, direct sources, referral sources, social sources, new customers,bounce rate, and returning customers exhibit positive skewness, implying a concentrationof values towards the higher end. This suggests that certain agri-food firms allocate signifi-cantly higher resources to advertising activities, customer acquisition, and engagement.Conversely, pages per visit and search sources display negative skewness, suggesting apreponderance of lower values, possibly indicating areas of improvement or optimization.Furthermore, kurtosis, which characterizes the shape of the distribution, provides addi-tional context. Leptokurtic distributions, observed in advertising costs, referral sources,new customers, and returning customers, suggest peaked distributions with heavier tails,reflecting concentrated expenditure and engagement levels. In contrast, platykurtic dis-tributions, as seen in social sources, search sources, bounce rate, pages per visit, and timeon site, indicate flatter distributions with lighter tails, illustrating greater variability inthese metrics across agri-food firms. These findings underscore the importance of consid-ering both skewness and kurtosis in understanding the distributional characteristics andpotential implications for advertising strategies within the agri-food industry.The correlation analysis of advertising costs (see Table 3) reveals its relationship withvarious variables. It is known that although paid ads do not directly affect organic websitesearch results, they indirectly impact other metrics that tend to cause a significant effect onSERP and SEO results [69]. We observe a positive correlation with social sources, directsources, time on site, and new customers, while it exhibits a negative correlation withreferral sources, search sources, bounce rate, returning customers, and pages per visit.The positive correlation with social sources suggests that allocating more resourcesto social media for promotion, results in higher overall advertising costs. The positivecorrelation with new customers implies that acquiring new customers positively relatesto overall advertising costs. Moreover, the negative relationship between the number ofreturning website customers and advertising costs suggests that retaining and reacquiringcustomers may require less investment in promotion and advertising. Conversely, attract-ing new customers is positively correlated with higher advertising expenditures. Therefore,agri-food firms should carefully balance their strategies to attract new customers whilenurturing existing relationships to maximize the effectiveness of their advertising expendi-tures. The research by Michel et al. [70] has a conclusion similar to previous findings; inan agro-industrial company, it seems that digital marketing has a moderately significant Sustainability 2024,16, 5889 18 of 25relationship with social media marketing and content marketing and a high and significantrelationship with customer acquisition.The negative correlation with search sources suggests that increasing expenditureon search advertising is linked to a decrease in total advertising costs. Enhancing searchengine presence is a critical component of digital marketing for agri-food enterprises. Byemploying SEO strategies, like keyword optimization, businesses can boost their website’svisibility and draw organic traffic from potential customers. Improving the overall websiteexperience is vital for retaining visitors and motivating them to explore the offerings further.This involves ensuring easy navigation, providing informative content, fast loading times,and a seamless user interface. Bhatnagar et al. [71] highlight that poor navigation impactswebsite design by prolonging visits and decreasing the likelihood of a purchase. Thenegative correlation with bounce rate implies that lower bounce rates are associated withhigher total advertising expenses. Furthermore, a rise in website abandonment rate is linkedto additional decreases in advertising expenses. Effectively managing and optimizingwebsite engagement to reduce abandonment rate can therefore lead to even greater savingsin advertising costs. The negative correlation with pages per visit indicates that an increasein the average number of pages per visit is linked to a decrease in total advertising costs.This highlights the significance of delivering an engaging user experience on the websiteto boost user interaction and potentially reduce advertising expenditures. Afterward, anexamination of the hypotheses, formulated in Section 2.4, is carried out and the resultsare interpreted.Based on the standardized coefficients and the p-value provided for the variables socialsources and search sources (see Table 4), it appears that both variables have a significantimpact on advertising costs. Therefore, the results support Hypothesis H1. Specifically, anincrease in the social sources variable is associated with an increase in total advertisingcosts, and an increase in the search sources variable is associated with a decrease in totaladvertising costs. The model explains approximately 68,9% of the variability in advertisingcosts, social sources, and the variability in advertising costs, search sources. Note that thep-value is statistically significant at the 0.05 level, suggesting that the relationship is likelyreal and not random. The Durbin–Watson statistic indicates that there is no correlationamong the residual deviations, indicating no autocorrelation in the model.There is a negative correlation between advertising costs and search sources, sug-gesting that allocating more resources to search engine optimization (SEO) and relatedstrategies may result in lower overall advertising expenses. Conversely, there is a positivecorrelation with social sources, indicating that increasing investment in social media forpromotion typically leads to higher advertising costs. This underscores the importance ofbalancing SEO efforts, such as keyword optimization and content enhancement, to improveorganic visibility without significantly increasing advertising expenditures. Effective SEOcan drive organic traffic and reduce reliance on paid advertising channels. On the otherhand, enhancing social media presence often involves expenses associated with contentcreation, paid promotions, and community engagement efforts. Agri-food enterprisesshould strategically manage both SEO and social media investments to optimize theirdigital marketing efforts while controlling advertising costs effectively.The standardized coefficient for bounce rate is−2.554, and the p-value associated withthis coefficient is 0.049 (see Table 5). This indicates that there is a statistically significantrelationship between advertising costs and the abandonment rate of websites. The negativestandardized coefficient suggests that an increase in the abandonment rate of websites isassociated with a further decrease in advertising costs. Additionally, the model explains ap-proximately 70.8% of the variability in advertising costs. Therefore, based on the provideddata, we can support Hypothesis H2, indicating that there is indeed a relationship betweenthe advertising costs of agri-food firms and the abandonment rate of their websites.The negative standardized coefficient and significant p-value indicate a strong relation-ship where higher abandonment rates on websites are associated with lower advertisingcosts for agri-food firms, highlighting the critical role of website engagement in marketing Sustainability 2024,16, 5889 19 of 25strategies. Higher abandonment rates typically signal that visitors do not find the websiteengaging or relevant. This underscores the importance for agri-food firms to focus onimproving website usability, content relevance, or targeting strategies to decrease bouncerates and potentially enhance advertising effectiveness.The negative standardized coefficients for both pages per visit and time on site (seeTable 6) indicate that an increase in these variables is associated with a decrease in ad-vertising costs. The model explains approximately 63.2% of the variability in advertisingcosts for pages per visit and for time on site. Both p-values are statistically significant atthe 0.05 level, indicating that the relationships are likely real and not due to chance. TheDurbin–Watson statistics for both variables are close to the ideal value of 2, suggesting nosignificant autocorrelation in the model residuals. Overall, based on these results, we canconclude that there is a significant relationship between the advertising costs of agri-foodfirms and both the pages per visit and time spent on site by customers on their websites,supporting Hypothesis H3. The negative standardized coefficients for both pages per visitand time on site indicate that an increase in these variables is associated with a decrease inadvertising costs. This suggests that as users engage more deeply with the website (viewingmore pages and spending more time), the costs associated with advertising decrease. Thiscould be due to more effective targeting or better quality of user engagement leading tolower necessary advertising expenditure.According to Table 5, the negative standardized coefficient for returning customerssuggests that an increase in the number of returning website customers is associatedwith a decrease in advertising costs. The model explains approximately 70.8% of thevariability in advertising costs. The p-value is statistically significant at the 0.05 level,indicating that the relationship is likely real and not due to chance. The Durbin–Watsonstatistic is close to the ideal value of two, suggesting no significant autocorrelation in themodel residuals. Therefore, based on these results, we can conclude that there is indeed asignificant relationship between the advertising costs of agri-food firms and the number oftheir returning website customers, supporting Hypothesis H4.The negative standardized coefficient and significant p-value highlight a robust re-lationship where a higher proportion of returning customers is associated with loweradvertising costs for agri-food firms. This underscores the strategic importance of customerloyalty and retention efforts in optimizing marketing expenditures and maximizing overallprofitability in the agri-food sector. Additionally, the highest R2value (approximately70.8%) suggests that returning customers and bounce rate are significant factors influencingadvertising costs in these firms. This robust correlation emphasizes the need for strategiesthat enhance the user experience on the website and improve the overall efficiency ofadvertising efforts in the agri-food sector.This article, after examining the relationship between digital marketing metrics andadvertising costs, aims to propose a digital marketing strategy. The strategy is intendedto result in a reduction in advertising costs. The two variables that were the focus of thesimulation are search sources and social sources. The selected pair of variables are crucialcomponents in formulating a comprehensive digital marketing strategy aimed at reducingadvertising costs. The statistical significance of the ANOVA analysis being 0.048, indicatesthat there is a low probability of obtaining these results by chance alone. This suggests thatthe relationship between the variables (social sources and search sources) as a whole modelis statistically significant. In practical terms, this means that the inclusion of these variablesin the analysis has a meaningful impact on understanding and predicting outcomes relatedto digital marketing strategies.Thus, five possible scenarios emerged, which are shown in Figure 2. These scenarioswere chosen strategically to highlight the potential for significant cost reductions, aligningwith this study’s policy recommendation. Specifically, this study proposes that agri-foodbusinesses should allocate their resources toward search sources rather than social sourcesto minimize promotional costs. This strategic direction underscores this research’s aim Sustainability 2024,16, 5889 20 of 25to optimize digital marketing investments for sustainability and profitability in the agri-food sector.In the first scenario, which is reflected in Figure 2a, an increase of 100% in the socialsources variable is carried out, and then, there is an increase in the advertising cost variableby 6%. Then, the second scenario is carried out, which is depicted in Figure 2b. In this case,a reduction in the social sources variable by 100% is carried out. This results in a reductionin the advertising cost variable by 13%. The third scenario is depicted in Figure 2c andconcerns the increase in the search source variable to 100%. A slight reduction in advertisingcosts by 1% is observed. Afterward, in the fourth scenario depicted in Figure 2d, a reductionin the search sources variable is attempted by 100%, which results in an increase in theadvertising cost variable by four percent. Ultimately, the fifth scenario is executed, entailinga simultaneous modification of both variables. Consequently, an endeavor is undertaken toaugment the search sources variable by 100% and diminish the social sources variable by100%. It is observed that the concurrent modification of the two upper variables results in anoteworthy decrease of 14% in the advertising cost variable, as depicted in Figure 2e.The scenarios are designed to demonstrate a marketing strategy that achieves a reduc-tion in advertising costs. After each variable has been examined separately, they are thencompared simultaneously. The simulation shows that the fifth scenario offers the best digi-tal marketing strategy for the reduction in advertising costs. The distribution of resourcesinto social media marketing frequently results in cost escalation, as illustrated by the initialsimulation scenario. This outcome arises from multiple factors. Firstly, heightened com-petition within social media platforms drives up advertising expenditures, as businessescompete for visibility and engagement. Secondly, the dynamic nature of social medianecessitates ongoing monitoring, analysis, and adjustment of marketing strategies, de-manding additional human and financial resources. Platforms such as Facebook prioritizeestablishing personal connections with users to cultivate customer loyalty [72], intensifyingthe need for enhanced resource allocation to maintain these relationships. While socialmedia enhances web traffic, it does not necessarily lead to significant increases in productorders and sales revenue [47]. Additionally, targeting specific audience segments on socialmedia platforms often demands investment in advanced tools and technologies, whichfurther drives up costs. Therefore, careful resource allocation is essential in social mediamarketing to mitigate the risk of excessive spending and ensure efficient use of resourcesacross different marketing channels. This caution is warranted because managerial actionsmay not always yield the desired effects. For instance, in the fourth scenario analyzed,reducing investment in search sources ultimately resulted in higher advertising costs.Before utilizing social media for business purposes, it is crucial to develop strategiestailored to the product and target audience. According to Kilgour et al. [73], effective socialmedia marketing requires aligning messages with target audiences and achieving robustcustomer engagement. This principle holds particular relevance in the agri-food sector,where the target audience may not be highly active on social media platforms. Given thedemographic characteristics of agri-food consumers and the diverse nature of agri-foodproducts, conveying specialized knowledge and tailored messaging through social mediacan be challenging. Therefore, allocating resources to social media marketing for agri-foodproducts without careful consideration of the target audience’s online behavior and prefer-ences may lead to inefficient resource allocation and suboptimal results. The simulationindicates that the most effective strategy for reducing advertising costs involves reducingsocial sources of advertising while increasing investment in search engine marketing. Thisis the business operation proposal that our research suggests.This perspective contrasts with that of Inegbedion et al. [74], who argue that lever-aging social media platforms like Instagram and WhatsApp significantly reduces costsand enhances marketing efficiency, thereby increasing turnover in agricultural productsin South-South Nigeria. As our research revealed, the use of social media should notbe preferred as a means of digital marketing; however, it is an excellent tool for com-munication. Madonna et al. [75] highlight that platforms such as Facebook, WhatsApp, Sustainability 2024,16, 5889 21 of 25and Instagram can facilitate community participation in agriculture by promoting socialinteraction, discussions, and consultation, ultimately enhancing product promotion andmarketing through e-commerce to achieve sustainable development goals. Thus, whilesocial media offers valuable opportunities for engagement and community building inagriculture, its role in direct digital marketing strategies should be carefully evaluatedbased on specific industry dynamics and audience preferences.The agriculture and agri-food sectors are increasingly emphasizing sustainability andtransparency in supply chains, driven by rapid industrialization, growing global fooddemand, and heightened concerns about food quality and safety. According to Manglaet al. [76], research identifies ten factors that influence sustainable development in theagri-food sector. Two of these factors, which are also pertinent to this study, include under-standing customer behavior and effectively managing costs. These factors underscore theimportance for companies and the food sector as a whole to comprehend customer behaviorthoroughly. This understanding enables the strategic allocation of financial resources to-wards initiatives that reduce operational costs and ultimately minimize inefficient resourceutilization, thereby promoting sustainable development.As concluded by Hidayati et al. [23], value-chain governance that integrates sustain-ability drivers provides a holistic approach to balancing social and environmental impactswith profit maximization. This approach opens up higher-value markets for smallholdersin developing nations within the agri-food sector. Businesses in this sector can derivesubstantial benefits from implementing digital marketing strategies that cater to the distinctbehaviors of digital consumers. These strategies enable them to reduce operational costs,optimize resource utilization, and customize their offerings, thereby facilitating the sale ofproducts at higher prices.5. ConclusionsThe scope of this article is to explore the paradoxical phenomenon, wherein agri-foodbusinesses sell their products at high prices and show profitability despite production costs.Specifically, it aims to examine how specific digital marketing measurement variables relateto advertising costs. Also, our research tries to propose a digital marketing strategy, capableof reducing advertising costs, thereby increasing profitability and contributing to efficientresource utilization and sustainable development. This research used the linear regressionmethod to extract statistically significant results as well as the correlation between digitalmarketing metrics and advertising costs. After studying the digital marketing metrics,this research focuses on the variable search sources and social sources to submit a digitalmarketing strategy proposal. Afterwards, a simulation is carried out using the FCM model,and five possible scenarios are analyzed. The conclusions of our research are as follows:•Advertising costs of agribusinesses are positively and statistically significantly relatedto social traffic sources.•Advertising costs of agribusinesses are negatively and statistically significantly relatedto search traffic sources, bounce rate, the number of returning website customers,pages per visit, and time on site customers spend on their websites.•Optimal Resource Allocation: Agri-food businesses can achieve cost efficiencies byprioritizing investments in search sources over social sources in their digital market-ing strategies.•Impact on Sustainability: Effective digital marketing strategies not only enhanceprofitability but also contribute to sustainable practices by reducing advertising costsand resource wastage.•Strategic Recommendation: This study suggests that agribusinesses should focus ontargeted digital marketing efforts tailored to search engine optimization (SEO) ratherthan social media platforms.As the global population increases, it is crucial to develop strategies that enhance foodproduction, minimize resource consumption, and reduce environmental impact, addressingthe significant issue of food loss and waste throughout the value chains [77]. Since 1987,

---